New levels of understanding: methods for revealing structural links to chronic disease

摘要

Research on the etiology and course of illness typically follows a biological model emphasizing the individual as the primary unit of analysis. The result has been a narrowed orientation that focuses a great deal of resources on individual-level interventions and much less attention to underlying factors that contribute to the prevalence of disease. Chronic diseases such as asthma, coronary disease, ulcers, diabetes, and cancers, however, have been found to be affected by multiple factors on various levels of analysis, defying attempts to 'individualize' risk and prevention. The study of structural factors that underlie health outcomes can reveal important information that normally remains undetected during clinical and epidemiological research on individual risk factors. Research that incorporates contextual variables of a physical, social, biological, temporal, or economic nature can help identify patterns of illness across space and time and factors that predispose a population to greater health disparities. This article reviews the theoretical basis and methods for improving the validity of ecological studies in addressing contextual variables. In particular, research design incorporating multiple levels of analysis and geographic information systems (GIS) is examined to address some of the challenges to incorporating community information. Community-based studies are described to illustrate this approach to ecological research. Finally, suggestions are offered for the further development of ecological and contextual research strategies in public health interests.